Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add PromptPartner/agentsmith --skill harness-doctorgit clone --depth 1 https://github.com/PromptPartner/agentsmithWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/promptpartner/agentsmith/harness-doctor)<a href="https://agentmods.dev/skills/promptpartner/agentsmith/harness-doctor"><img src="https://agentmods.dev/badge/skills/promptpartner/agentsmith/harness-doctor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/promptpartner/agentsmith/harness-doctor"><img src="https://agentmods.dev/badge/skills/promptpartner/agentsmith/harness-doctor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Rogue Agent · line 12 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00074 | $0.00743 |
| Opus 5 | $0.00037 | $0.00371 |
| Sonnet 5 | $0.00015 | $0.00149 |
| Haiku 4.5 | $0.00007 | $0.00074 |
Grade A, and why
harness-doctor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness doctor — is the harness healthy?
Confirm the harness is wired correctly, in plain language, with a concrete fix per finding.
When this fires
"is my harness set up right / healthy?" / "harness doctor" / after an install or a --self-update.
Identify the target
Use the agent named by the operator or the explicit AGENTSMITH_AGENT value. Never infer the active
agent from this skill's install path: portable canonical content may be copied into a required
client adapter such as .claude/skills. Without an explicit agent, run the common project checks
and label agent-specific capabilities as unknown.
Fast path — if the Agentsmith CLI is available
Run agentsmith doctor, adding --agent <id[,id...]> when the target is known. If the command is
not on PATH, use .agentsmith/agentsmith on macOS/Linux or .agentsmith\\agentsmith.cmd on Windows. Report instructions,
actual safety, skills, MCP/config, hooks/scanner, and runtime helpers separately. For instructions,
report the effective global/project/nested chain, fingerprints, managed core metadata, and
combined/duplicate token estimates. An unsupported optional capability does not make instruction
discovery unhealthy, and a duplicate core is a warning rather than permission to rewrite it.
Fallback — self-contained project checks
Check each; report pass, or a one-line fix:
- Rules: canonical
AGENTS.mdis present and contains the AgentSmith BEGIN/END managed block. When a target requires a generated instruction adapter such asCLAUDE.md, compare its managed content withAGENTS.md. Fix: update the intended agent through the Agentsmith installer. - Configuration: validate only the declared target's native configuration. For Claude, check
.claude/settings.jsonand/orsettings.local.json. For Codex, validate.codex/config.tomlwhen present and the managedconfig.tomlunderCODEX_HOME(or the user's.codexdirectory when that variable is unset). Never treat one client's configuration as proof of another's. - Skills and hooks: inspect the shared
.agents/skillspack first, then any genuinely required client adapter and the target's declared hook integration. Never report hook parity for a client that only discovers instructions or skills. .harness/verify.confexists AND has a real phase — not just thesanity ::placeholder. Fix: edit it to list this project's real checks.- Runtime helpers:
agentsmith verify --listsucceeds without requiring Bash or WSL. - Leanness: use the compatibility report's static-context measurement for canonical
AGENTS.md, and report the line/token budget. Over budget → move prose into a skill or doc, not morecore/.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago Changed · +1 lines ac2db7a7a437
- 9d ago First seen · 49 lines · 74 tokens per session scan A 818a6188df2c
harness-doctor is a skill published in the GitHub repository PromptPartner/agentsmith (185 stars, last pushed 4d ago), licensed MIT. It adds 74 tokens to every session and 743 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
mobile-performance-react-native
React Native performance profiling, optimization, and monitoring - JS/UI thread analysis, re-render prevention, list optimization, image performance, bundle size, startup time, memory leaks, React Compiler, New Architecture benefits.
api-observability-axiom-pino-sentry
Pino logging, Sentry error tracking, Axiom - structured logging with correlation IDs, error boundaries, performance monitoring, alerting.
api-observability-setup-axiom-pino-sentry
Pino, Axiom, Sentry installation - one-time project setup for logging and error tracking with source maps upload.
web-error-handling-error-boundaries
Error boundary patterns, fallback UI, reset/retry, react-error-boundary library, React 19 createRoot error hooks.
backpropagation
Trace runtime bugs back to spec gaps — identify missing acceptance criteria, update specs, generate regression tests, and detect patterns.
oo-airbrake
Airbrake (airbrake.io). Use this skill for ANY Airbrake request — reading, creating, and updating data. Whenever a task involves Airbrake, use this skill instead of calling the API directly.